Smalltalk Tidbits, Industry Rants

One of the things I see being floated around the blogosphere is how woefully unprepared for an avian flu outbreak the country (and the world) are. Hmm - The last time we had a real pandemic was 1918.

What was going on then? Well, Europe was engulfed in WWI. China was slowly crumbling into anarchy at the tail end of the period of European incursions. India was still a British colony, and had far less infrastructure than it has now. Southeast Asia outside of China? A complete backwater, with virtually no infrastructure. The state of medicine was primitive in comparison - there were no flue vaccines, period.

And yet, in the midst of the disaster that was WWI, the flu spread across the world - yes, killing millions - but without significant upheavel in areas that weren't already in upheavel (like, say, the Western Front in France). There weren't riots in London, or in Washington, or in New York. People dealt with the problem.

So to all those who are preaching fear and gloom? Grab yourselves a cup of historical perspective and calm the heck down.

This tutorial has been around for a while (the date on it is 2002), but I just noticed it. Looks like a good resource for Java developers who would like to take a look at Smalltalk, and relate it to what they already know.

Sometimes, people using BottomFeeder with lots of subscriptions (or with a large cache setting) report crashes due to an "out of memory" error. Here's why - in settings, there are two values:

• softMemoryCeiling
• absoluteMemoryCeiling

Bf tries to keep the application within that range. If you have "too many" feeds for that range, it can run into problems. That's something I need to look at, but in the meantime, if you have errors like that - up those values. You can just edit the file btfSettings.ini and up them before you start Bf.

Steve Rubel takes some exception to a statement from Jim Lanzone of Ask Jeeves:

Jim Lanzone from Ask Jeeves gave some interesting data about feed subscribers at this week's Web 2.0 conference. He said that 14,363 blogs have at least 50 subscribers on Bloglines. Lanzone calls these blogs that “really really matter.” Uh, don't they all matter? Jim, this is 1.0 thinking. This is not a numbers game. A blog with two readers can “matter” if those two readers are people like Robert Scoble and Doc Searls. God help Bloglines.

Here's the thing though - if a site only has two readers, what's the actual liklihood that either of them will be a connector? Not high, I'd warrant. The number of readers may be a sub-optimal measure of relevance, but it's as good a measure as any that we have. Not to mention this - if a top connector is one of the handful of readers, it won't stay that way - as soon as that connector points to the blog, readership will go up.

Dave Winer makes some telling observations about Google's new reader:

I tried the Google news reader again, this morning, after it had loaded all my feeds (it seems to take quite a few hours to do that).This is the second blog-related product they've come out with recently that appears not to have been touched by human beings before it was introduced to the world (the other was the ridiculous blog search). I think they need to start using their own stuff before releasing it. And maybe look at the competition for ideas.

The emphasis above was added by me - I think that Dave hits it right on the nose there. BottomFeeder may not be perfect, but I use it myself everyday - eating your own dogfood is critical with customer facing software. It would be very interesting to see just how much user testing this tool got. As I said earlier, I don't think people like me are their primary target. Having said that, I'm not sure which target would actually find this tool usable right now.

BottomFeeder downloads have dropped back toward the older level - right at 402 per day this week. Here's the distribution:

Platform	BottomFeeder Downloads
Windows	830
HPUX	510
Mac 8/9	388
Sources	354
Mac X	220
Linux x86	173
Update	135
CE ARM	108
Linux Sparc	23
Solaris	20
Windows98/ME	18
AIX	14
SGI	10
Linux PPC	6
Source Script	3
ADUX	2

Interestingly, the drop was in Windows downloads - everything else has stayed pretty much the same. Maybe the spike was simply due to the new version. So moving on, here's the report on HTML page accesses:

Tool	Percentage of Accesses
Mozilla	52.5%
Internet Explorer	36.5%
Other	5.2%
MSN Bot	3.4%
Google Bot	2.4%

Looks pretty normal - no real changes there. Let's have a look at the RSS distribution:

Tool	Percentage of Accesses
Mozilla	26%
BottomFeeder	17.6%
Net News Wire	9.8%
Other	13.1%
Feed Demon	6.1%
Internet Explorer	4.3%
Safari RSS	4.3%
Planet Smalltalk	3.3%
NewsGator	2.9%
Magpie	2.7%
SharpReader	2.6%
BlogLines	2.2%
Feed Reader	1.7%
RSS Bandit	1.4%
BlogSearch	1.3%
Liferea	1%
RSSReader	1%
Google Bot	1%
JetBrains	1%
Feed Tagger	1%

Interesting that FeedDemon usage seems to be rising since the Newsgator acquisition.

I'm in the field, so it's worth having a look at Google's new reader. I haven't done more than export my feeds out of BottomFeeder and start importing, but here's my first impression - they really, really need to give the user some kind of feedback during the import process. I sent it over 300 feeds in my OPML file, and all there is for feedback is a text box that says "Your subscriptions are being imported..."

No animation, no progress report. Heck, for all I know it's locked up. I'll have more to say once I see what happens next.

Update: Well, with no sign from the server that it ever finished importing, I simply refreshed the page. It apparently imported everything, but I have no idea as to how long it took. The interface does feel slow and clunky, as others have reported.

I think the main take away I have is that people like me are not Google's primary target. I follow hundreds of feeds - there are people out there who follow more. I suspect that Google's target reader keeps track of far fewer sources - 5 to 10, maybe. The two pane presentation style probably works for that kind of list, but I can't see me using it - BottomFeeder makes it far easier for me to spot the new stuff I care about quickly.

I have to say that I was pleased to see the Red Sox go three and out. However, it looks like the Yankees are trying hard not to make it to the LCS either - they gave up a ton of runs in a game they should have won tonight - and that's after the gift wrapping they did two days ago. Sigh.

Google has jumped into the online aggregator game. That's going to impact BlogLines, I'm sure. It also makes Newsgator and Bloglines bait for Yahoo, I think. This move is going to drive more consolidation, that's for sure.

That's what the powers that be (and 4 of the playoff teams) have to be singing right about now. Have a look at this forecast for the next few days - the temps are a little lower for Boston, but the graphic below is mostly the same for NYC and Boston over the next few days:

If that forecast holds up, things could get downright ugly in scheduling land.

I ran across two stories just now that are just amazing. First, let's have a look at this Boing Boing story about a web "intrusion". A UK guy donates money to a disaster relief, and then thinks maybe he's gotten scammed by a phishing thing. So - using only his browser and the keyboard, he tries a couple of things - and gets convicted of hacking for his trouble:

Stephen de Vries sez, "The details of this case are important to understand exactly how absurd the verdict was. What Daniel actually did to 'knock on the door' was to insert a ../../../ character sequence into the web address and a single quote into the credit card field - THROUGH HIS BROWSER. He did not use any attack 'tools' or 'probes' other than Internet Explorer. Furthermore, typing these sequences into a browser does not an attack make - it only proves that a website may be vulnerable. It takes a hell of a lot more effort to actually gain any form of unauthorized access to the site. Daniel did none of this, he only typed the sequences and watched the responses

I wouldn't put it past a US judge to have no clue either - it's all too easy for a well funded lawyer to paint scare quotes around this sort of thing, and exploit the lack of knowledge of a judge and/or jury. Next, I ran across some more stupidity from the RIAA (what a shocker):

The record industry may next aim its legal guns at satellite radio over a dispute involving new portable players that let listeners record and store songs, an analyst and industry sources said Wednesday

I guess the tape recorder is something these clowns haven't heard of. Hey stupid execs: I have a news flash for you - I was recording songs off the radio back in the late 60's, when I was under 10. It's not like this is a new thing, you half-wits.

I swear, the need for a massive cluestick gets bigger every day...

There's some worry from the NY sportswriters about the effects of a rainout of today's Yankees game - it will mean the loss of a travel day if the series goes to five days. Well, if you look at where this storm is headed, you'll notice something - it's entirely probable that the Red Sox/White Sox game will be washed out as well (and in their case, possibly Saturday's game as well). Which means it's a lot of worry about nothing :)

The Register has the goods on the purchase of Weblogs.com from Dave Winer - this is the ping service, not the content provider that AOL bought this week. What it looks like we'll see is some form of QOS added for a fee:

VeriSign has promised that the basic ping service will remain free, but "over time, we plan to offer value-added services... in much the same way companies like Yahoo! provide basic email services for free, and offer premium 'upgrades' for a fee."

The company then went on to discuss the problem of splogs and what could be done with them. "This problem is fraught with many of the same problems that plague the email world in its struggle against spam: Who is the source? What is the content about? Is it a copy?... However, at the infrastructure level, very little is currently being done, and there are remedies that can be deployed that will provide significant, if not thorough relief."

I can't see any way to justify the cost of running a ping service unless you charge for QOS somehow - I'd guess that the splog filtering will be a pay service.

You think Microsoft hasn't had to adapt much over the last few years? Have a look at this post - a challenge to Mini-Microsoft from an MS VP. There's a public airing of difficulties that we (outsiders) wouldn't have seen a few years ago.

I told my wife that Word sucked every bit as much as WordPerfect, just differently. She's been using the product for a few months now, and is every bit as pleased with it as she is with Outlook - and bear in mind, her point of comparison is with the default "productivity" apps that Sun ships on their Unix workstations (or more precisely, the ones they shipped 2-3 years ago).

A rather low bar, and Word consistently manages to not make it. Last night, she was putting together a checklist of items for a girl scout event - she's one of a group of Mom's helping co-run my daughter's girl scout troop this year. It's not a complex checklist - some text, a few bullets, more text, a few more bullets. After I got past the swearing, she told me that Word was an AD product (as opposed to AI).

What's that mean? Artificial Dominatrix. It puts the bullets where it thinks they belong, and if you try to argue with it, you will be punished. The Word team at MS needs to be taken out behind the woodshed, and beaten severely with the cluestick. The menus/toolbars/ribbon area is so not the problem. The basic "I'll guess what you are trying to do and then preempt you" behavior is.

I have somewhat fond memories of Word for Windows 2.0. Each release since then has sucked more and more - and sadly, tools like WordPerfect and OpenOffice have seen the suckage and copied it.

Scoble isn't happy with the results he gets from the search engines. He gives an example of searching for the term HDTV in all the major search engines, and then focuses on Google's results for his issue:

But I'm in a different role. I want to buy one.

So, let's just focus in on Google since that's the hot search engine of the moment. First link: an introduction. I don't need that. I already had an introduction. Second link: how HDTV works. I don't care. Next. Third link: an info site about stations and some product comparisons. Hmmm, maybe useful later, but I'm looking for something else right now. Fourth link: Amazon.com. Huh? I'm not ready to buy yet. I wanna know what's available. It predicted I was in a different role. Fifth link: a magazine site. OK, it's clear the search engine isn't going to give me what I want, so I'll probably go off and read that site for an hour and come back. Sixth link: an ATI card? I'll have to put that on my gift list too. Seventh link: HDTV Buyer site. News and info. Another site I'll have to go and check out later. And on and on it goes.

The trouble is, there's absolutely not context available to the back end for that kind of search. It's 4 letters. I really don't know how anyone is going to satisfy Robert's request - he wants pre-sale info first, but someone else might well want "what is this HDTV thing I keep hearing about?". The thing to remember is, not everyone tracks technology like Scoble does. The other thing to remember is, how contextual do you expect a set of results for a 4 letter search term to be?

I suppose the engine could query you, but that adds clutter, and I suspect that usability testing would tell you that people get irritated by that. Further down, he writes this:

We know this can be done. Why? Cause Google did it for Seattle Hotels. Here's the Google result for Seattle Hotels. They make a nice little list of all the hotels available and even give you one of those Google Maps. MSN Search has the exact same thing. Yahoo goes even further. They have pictures and ratings!!

So, why can't they do this for HDTVs? Of course they can. It just hasn't gotten onto the dev list of any of the major engines yet. Yet.

Well, what he's missing is the extra context. If I type HDTV in, I've provided no extra context - no information on whether I need a definition, or information on buying, or what have you. It's a crap shoot. Seattle Hotels has that extra context - not only are you interested in hotels, but you are specifically interested in Hotels in Seattle. The difference between the two result sets is all about the amount of context provided.

Look at it this way - people are way smarter than search engines. If a guy stops you in the street and says one word - HDTV - in a questioning fashion - what are you going to think he wants? And the person won't be someone you know. The more context you provide, the better the answers. The less context you provide, the worse the answers will be.

JavaLobby discovers that Java has gotten baroque:

It took me about a week to get up to speed on .Net and was able to create a fairly decent functional application with it. For those of you who are saying "Yeah, but you can do the exact same thing in Java! And it's portable." To that, I would say the same thing I said almost a decade ago; If it’s easier to use, people will switch over to the path of least resistance. Everything you can do in Java and .Net can be done in assembly, but why don’t we? Why should I spend 6 months on a Java project when I can cut that down to 4 in .Net?

Or 2-3 months in Smalltalk. Come on in, the water is warm :)

I can read Hani's thoughts on product evangelists. Slight warning - he doesn't hold off on the obscenity - but if you're ok with that, he's hilarious.

NPR reported on something fascinating:

NPR program “All Things Considered” reported yesterday that scientists are considering how the recent accidental spread of a plague in massively multiplayer game World of Warcraft might help them develop model responses to real-world disease outbreaks.

Of course, there won't be any disease resistant orcs to deal with in the real world.

For those of you with a twisted sense of humor, you might want to try counterscript out on the next telemarketer that calls. I have to admit, it sounds amusing :)

We (STIC) have been discussing next year's Smalltalk Solutions. We would like to attract people outside the existing community, so we had an idea - why not have Smalltalk Solutions as a set of tracks at a larger industry trade show?

So here's the thought - next year there's a LinuxWorld in Canada (Toronto, specifically). We've talked to reps from that show, and they are interested in having us there. Any thoughts on this? Positive, negative, neutral? Let me know.

Dana VanDen Heuvel uses his morning coffee experience to explain how good customer service generates return business. It's something that far too few companies pay attention to.

As much as I prefer the Spicy chicken sandwich at Wendy's, this is the same thing that drives me to McDonald's when I go for fast food. Wendy's service is just too variable - I've seen bad McDonald's service, but it's much rarer in my experience.

Well. back on August 9th, Dave Winer said this:

Big software companies, or BigCo's for short, just can't leave well enough alone. They always try to f*** with technology they didn't invent, for a lot of complicated reasons I've spent an entire career trying to understand. It's not powerful, or interesting -- it's childish and self-defeating, but it's evident in both BigCo strategies for RSS.

Now comes word from Kottke that he's sold weblogs.com to Verisign. Hmm. What was that again about big companies?

Ever have that "oh, I ate too much" feeling? I bet it was nothing like what this python experienced after trying to swallow a gator. Ouch.

Warning - if you click through to the full size picture, be ready for a nasty image.

I ran out of business cards a little while ago, and I just got around to going to our internal form for re-order. That's when I realized that even something as simple as business card ordering hadn't caught up with current trends. By default, the web address of the company (Cincom) will appear in the lower left of the card. That's fine - it should. However, I also wanted the address of my blog there, under it. I was able to work with the order folks to make that happen, fortunately.

Sometimes, it's the small stuff.

Schwartz reaches for meaning, and fails. I guess he noticed that the meaningless PR event with Google was, well - meaningless, and tried to inject some meaning into it:

Two of the internet's most valuable brands have clearly achieved that status [distributor level importance]: Java and Google. Could you imagine a PC that couldn't access Java services? Or how about a browser that couldn't get to Google? My view, either would be a tough sell. Other programming, such as Macromedia Flash, Firefox and OpenOffice are in the same league - along with services such as Yahoo.com, eBay, or AOL.com. The world of network services is enormously competitive, and it's driving enormous innovation (and quite a few deals hearkening back to the early days of the internet). Much of this next wave of innovation is referred to as Web 2.0 - the convergence of technology and services that underlie the Participation Age.

The most telling aspect to that nonsense is the "hearkening back to the early days of the internet" - citing deals that were made by Netscape early on don't really help make his point, IMHO. And a PC without Java services in the same league as one that can't access Google? Java is hugely popular, yes - but a force on the client browser side, it's not. If it disappeared from the browser tomorrow, few people would notice or care. To take Google away, you would have to remove the network. That would get noticed.

The bottom line for Sun - they just aren't that relevant at the "movers and shakers" level. It was nice of Google to take pity on them and make a nothing-burger announcement, but it smells of pity.

Men left on base all night, and three errors. Sheesh. Back to New York, tied 1-1.

Well - the White Sox have taken two straight from the Red Sox, and tonight's loss had to hurt worse than last night's - The Red Sox were up 4-0 in the 5th. Two runs came in, and the White Sox had a man on first. A soft roller was hit to second base - and Graffanino did his best Bill Buckner imitation - he started to second before the ball was there, and it went into right field. Instead of the inning ending, there were men on 1st and 3rd with one out.

Next batter - 3 run shot. The White Sox held that lead the rest of the way through, and they are now up 2 games to none. If the Red Sox lose this series, Graffanino will be remembered in much the same way as Buckner is - for good or ill.

Mark Bernstein spotted this exercise in editing - take scenes from the Original "The Shining", and make a trailer that portrays it as a Romantic Comedy:

Here's the winner of an interesting contest: take a Hollywood film and cut a new trailer for it -- in a completely different genre. Robert Ryang won with a fresh cut of The Shining as a romantic comedy.

This was so convincing that I was wondering to myself, 'How could I have missed this unsual Jack Nicholson movie? And wasn't there a Steven King novel with the same title?'

It's interesting all right. Based on how dark the original movie was, it's really interesting. Mildly disturbing as well :)

The next largish task I have in mind for BottomFeeder is an interface to the Newsgator online API. Fortunately, the grunt work of generating the basic interfaces was handled by the WS* wizard in VisualWorks; now comes the hard part - hooking one end up to Bf, and the other end up to Newsgator. Among the other things I'm sure I'll have to deal with is the post id problem. For feeds that don't have GUIDs (and there are plenty of them), what does Newsgator expect? I'll have to read up.

Update: If anyone's interested, I've posted the code to the public Store as a Bundle - Newsgator-API.

Wow, there's a new tropical storm named Tammy out there - fortunately, it formed so close to the coast of Florida that it will land before it becomes dangerous. Looking here, that leaves us with Vince and Wilma before we exhaust the standard names and go into the Greek alphabet...

Yesterday, Google and Sun had a joint announcement of sorts, where they tried to pass of trivia as world shaking news. So I guess Google felt a need to say something more, in order to spice up the non-announcement:

Google boss Eric Schmidt worked at Sun Microsystems for 14 years, so it's no surprise he shares a vision for the future of computers and networks with Sun boss Scott McNealy.

And it's no surprise that future doesn't include a place for Microsoft.

Well, actions speak way, way louder than words. Let's look at the places where Google has reached out past the browser, or into browser add-ons - the toolbar, Google Earth, desktop search. Hmmm - there's a pretty Windows-centric aspect to all of those initiatives, isn't there? Heck, until recently, the toolbar was even IE specific.

If Google were actually interested in a future "sans Microsoft", you would think that their client tools would support the Mac and Linux, at least (and it says something about Sun's desperation that they were willing to appear with a vendor who will support Solaris approximately never).

The Blog Ride thinks that dynamic languages provide power that most people just can't handle:

Folks, we need to realize something: all this "expressiveness" is like putting craftsman's tools in your hands; in the hands of a master craftsman, amazing things can result, but in anybody else's hands, it's putting a loaded gun into the hands of a child. YOU may be good enough to be disciplined enough to keep the rules of your types in your head when programming with Ruby, but are all of the programmers on your team equally gifted? Are all of the programmers that will follow you so gifted?

I think David Buck's latest post explains what's at stake quite nicely. To my mind, it's all about just how much productivity you are willing to sacrifice in the name of presumed safety.

It doesn't get much better than Mussina leading the Yankees to a 4-2 first game win, while the White Sox crushed the Red Sox 14-2.

Travis is looking for Smalltalk blogs. Of course, the largest aggregation is right here, on our site. You can subscribe to the whole thing via the master feed. Here's what else I've run across that's worth subscribing to (Specifically Smalltalk):

• Blaine Buxton
• ESUG Info
• Learning Seaside
• Squeak People
• Vince Foley
• Planet Smalltalk

I'm not the only one who thinks it's dumb. Go read this, and then ponder how much less working space you'll have with Office 12 and the absurdly enormous ribbon/toolbar area. I guess someone at MS liked the huge amount of wasted space at the top of Outlook, and thought "we need more of that around here".

Sadly, WordPerfect still sucks, and Open Office sucks even worse. sigh.

There was a lot of anticipation about the Sun/Google announcement this morning, and then it amounted to nothing. Whoo Hoo - you can download the toolbar when you come to Sun's site. Wow. Om Malik is less impressed than I am, and given how down on Sun I am, that's saying something :)

This morning's rumors are true - NewsGator has acquired NetNewsWire. The consolidation has begun.

John Dvorak makes some points on advertising's problems - mass market advertising isn't contextual, and an increasingly large share of the market is able to avoid it now. No one has ever really like mass advertising - it's just that now, we have options:

The advertisers should have gotten a clue when, almost out of the blue, full-season episode lineups of TV shows began to sell on DVD. You could watch without commercials. People would actually pay to avoid these commercials. Hello! Do you think you have a problem yet?

I went to the theater the other day to see the trendy film Serenity (I prefer watching movies at home on an HD projector) and had to endure about 20 minutes of solid commercials before it began. So now I pay money to watch commercials? It's offensive.

Friends of ours just started watching "Smallville" - by buying the season DVD's. They avoid ads, and they catch up quickly. I only started watching "Buffy" during season 5 - I caught up via the FX repeats captured on my ReplayTV, without the commercials.

The problem is, where does that leave the mainstream TV industry? Right now, they build shows one step removed from the actual audience - they target the advertisers. However, with the ability to dodge ads on the rise, that model is doomed. I rather suspect that over the air TV (with ads) is doomed - but I don't see targeted ads being the answer. Instead, I think we'll see subscription TV. Me? If I knew I was going to see content from Joss Whedon, for instance, I'd sign up.

James Governor wants some interop between competing IDE's (by which he means NetBeans and Eclipse) - at least at the plugin level. He asks:

But couldn't we learn from the wireless providers and at least have a roaming agreement?

Well, that assumes that the infrastructure underpinning the two are similar enough to allow for plugins that work in both. I seriously doubt that's the case, nor will it be anytime soon.

Anything can happen in sports, but here's my read for round one - The Red Sox should take Chicago. Their hitters are too damn good, and the White Sox don't have the kind of pitching staff (like, say, Houston has) to just shut them down. What about Yankees/Angels? That's tougher. The Angels won the season series, and they have home field advantage. The Yankees have something to prove though, and that might make a difference. Might. The big problem for the Yankees is what happens when the starter doesn't have his stuff - the middle relief staff sucks eggs. So this hinges on the starters - if the starters are good, the Yankees go to the ALCS. If they falter, the Angels go.

If we see a Yankees/Red Sox rematch in the ALCS, anything could happen. Both teams can hit, and both teams have nothing in middle relief. The Yankees might have an edge in the late game with Gordon and Rivera, but that assumes that the late game is close. If this rematch takes place, watch for fireworks.

I don't follow the NL that closely, but everyone is picking the Astros - with the Clemens/Pettite combination, they look a lot like the 2001 Diamondbacks. I expect them to make it to the series - and then woe betide whoever has to face them.

Well, there's a rumor floating that NewsGator is going to acquire NetNewsWire (the most popular Mac aggregator - by some accounts, the most popular aggregator period). I figured consolidation was coming to the aggregator space - it's a field that a lot of people (and money) are interested in, and there are tons of tools in it (witness any of my log reports - just look at the aggregator tool distribution).

Hat tip Scoble.

One of the things missed in last week's kerfuffle over OPML is how the semantic part of the semantic web is going to get driven. Let's look at two of the proponents of OPML organization for an example of the problem - and no,I don't mean this as a slam. First, Scoble, who was excited over the OPML work started by Dave Winer. Scoble's blog doesn't even have categories. Which means that, from a semantic web standpoint, his posts are lost in the ether. It's not that categories are the end-all, be-all of organization - far from it. They are kind of the bare minimum required, actually.

Categorization is hard, actually - where does a given post fit? Sure, some blog systems allow for multiple categories, but - if people won't bother to assign one, what makes you think that they'll start assigning 3 or 4?

Next, we have Dave Winer - who's posts not only don't have categories, they don't even (typically) have titles. His posts have even less of a chance of being organized in any sem-web kind of fashion. Like Scoble's posts, they end up not having any of the metadata necessary to hang additional tools off of.

Which gets us to the root problem of RDF, which isn't complexity. It's a lack of user supplied metadata. The OPML categorization scheme that is being proposed breaks up on the same set of rocks. Scoble and Winer are hardly the only people blogging without extra metadata - they are just too of the most well read bloggers doing so, and as such, they make for a good example. Let's say we wanted to use RDF (or the posited OPML scheme) to created an organized taxonomy of posts (say, over the last few weeks) of either of their blogs. We can't. Not unless we first feed the raw data of each post through a text processor which will create tag data of its own. That might be accurate; heck, I don't really know what the state of technology is in the area - but based on some of the results that come back from search engines, I'd say that it's less impressive than we'd like for for that kind of categorization.

Ultimately, the biggest blockage on the road to the semantic web is us - and I see no long term solution to that social problem.

Patrick Logan points to a debate over CORBA and WS* between Roger Sessions and Terry Coatta over at ACM Queue. He points out some fun stuff, like this exchange here:

RS If you’re using the J2EE standards such as RMI (remote method invocation) over IIOP (Internet Inter-ORB Protocol), you are primarily going to be doing that within a single vendor’s system, such as a WebSphere system. If you’re going from a WebSphere system to a WebLogic system, your best shot at interoperability is through Web services. Why? Because you’re crossing a technology boundary.

TC You’re claiming that RMI over IIOP doesn’t actually work?

RS It doesn’t work for interoperability across technology boundaries.

TC

There seem to be people out there getting it to work. Certainly, back in the days when I worked with CORBA there was no problem having different vendors’ ORBs (object request brokers) interoperate with one another. We used three or four of them at Open Text and had no difficulty at all with those environments interoperating with one another.

RS As long as you’re going CORBA to CORBA, it works fine. But not when you are trying to get a CORBA system to work with a non-CORBA system.

TC But going from WebSphere to one of the other EJB vendors (e.g., WebLogic) in the CORBA space, there were probably five or six different major ORB vendors floating around, not to mention a couple of open source efforts, and all of those interoperated really well with one another.

RS CORBA to CORBA. They’re all running on the same basic core of CORBA technology. The difference between that and Web services is that for Web services, unlike CORBA, there is no assumption whatsoever about what the underlying technology is.

Yes, one CORBA end point only works with another CORBA end point. Does that mean that Sessions believes that WS* works differently? Can I send a plain text (as opposed to XML) document via FTP to a WS* server listening via HTTP? What point is Sessions trying to make here? WS* is no more or less interoperable than CORBA was. The difference? The fact that most web services run via HTTP (and thus, via port 80) - so they can move between firewalls. With CORBA, you have to get IT (or IT security) to open up a port in the firewall (on both ends). The usage of HTTP as a transport gets around that problem.

That's really the only difference between the two at the 50,000 foot level. Lower down, of course, WS* manages to be even more complex (as Patrick points out). Managing to be more complex than CORBA is a rare achievement.

Dana VanDen Heuvel is not amused by totally locked down internet services. I'm not either - I've run across such madness at a few hotels, and at some corporate sites. Very, very unfriendly...

Hmm - Tom Yager imagines that he's "been on his game" lately:

I'm not just bored with the present, I'm uncomfortable there. I drop in on now every so often to do things like pick up my mail, write reviews, and make sure the pipes didn't freeze last winter, but before the neighbors get a chance to screw with my mind by asking me what I've been up to, I'm back to that windowless place where, for all I know, the horizons to which I cast my gaze are already behind me.

When I'm on my game, as I have been lately, I can haul my horizons with me when I travel on business.

Back here in the real world, let's look at some examples of Yager's fearless predictive powers:

• No, Apple won't go to intel.
• The big problem in IT is those blasted users
• iTanic is the future
• JIT? Who needs that?

If Yager were any more detached from the IT sector he'd be orbiting Neptune. With his track record, he belongs at my favorite analyst group.

I should do a screencast on this, but I don't have time for one today. This is the other side of the client demo I showed off a week or so ago. Let's say you want to expose some Smalltalk service as a web service - how does that work? Well, let's posit a class like this one, with a simple API. These are all reduced size images; you can click through any of them for a full size image:

So what we have is a simple class, with a method that answers a Timestamp. What we want to do is expose that as a service. So, we grab the wizard from the launcher after loading the WSDL Wizard support, and select the appropriate option:

Then, we select the class we want to expose, and select the method(s) (via the list of protocols) that we want to make available:

Next, we have to define the service interface - Smalltalk is dynamically typed, but not everything in the world is - so we have to define the type information that this API will expose. The wizard supports defining complex data types, but this example posits a simple one - Timestamp:

Next, we tell the wizard what to call the class that will support the actual server API, and what package to shove it into:

Next, we tell it to generate a client for us:

With that done, the wizard hands us some code suitable for use as a simple test case, in a workspace:

Last, we want it to generate some WSDL, so that other tools (or even another Smalltalk image) can talk to this service:

Once we've done all that, you should have a client and server suitable for testing - as seen below:

And that's pretty much it - it works the same way for more complex cases. All in all, pretty easy to work with.

ESPN reports that the Yankees are unhappy with the way the Angels played their last game, when very little was on the line for them:

The New York Yankees will be on the road for Game 1 of their Division Series against the Los Angeles Angels. And the Bronx Bombers are crying foul about it, directing their ire against former Yankees manager Buck Showalter.

Showalter removed starters Michael Young, Mark Teixeira and Hank Blalock for pinch-runners in the third inning against the Angels on Sunday. The Texas Rangers led 4-1 after three innings, but the Angels rallied for a 7-4 victory and earned home-field advantage for the first round against the Yankees with the victory.

Hmm - seems to me that Torre decided to go with a sub-par pitcher instead of Mussina, once the AL East was locked up. So what the heck does he have to be angry at, other than the fact that his team went out and phoned it in for the last game of the regular season?

Here's something that everyone (including me) who's a product evangelist needs to keep in mind - the actual economics end up mattering. The link is to RealTechNews, where they examine the decision making process for hybrid cars (like the Prius):

“White first looked at trading in his Subaru for a Prius, and found that at roughly $3 per gallon for gas, he wouldn’t recover his financing costs. Joe figured that at his annual mileage, he’d save about $746 a year in fuel costs, but it would take too long to recover the premium he’d pay for the hybrid.

“Next he looked at the hypothetical situation of someone without a car looking to buy either a Honda Civic or the Prius. In this case, the fuel savings were roughly $506 per year, versus a purchase price difference of about $8,000. Without even considering cost-of-money issues, it would take nearly 16 years just to break even.” Source: Autoblog

Costs/Benefits analysis is something we all do implicitly. We don't always get the facts right (leading to seemingly odd decisions), but we still do it. In the example above, we see that hybrids aren't going to sell well until fuel gets a lot more expensive, or the premium goes down.

There's a lesson there for a lot of us.

Douglas explains the halting problem in terms anyone can follow.

That was fast. I noted that Blogniscient didn't have syndication support earlier today, and bam - I just got email indicating that they do - on the main tech and political pages, at least. Here's the tech feed.

Now that's fast turn-around

Remember the whole "Dell Hell" thing that caught Dell with their collective pants down? It looks like Apple is a whole lot smarter than that:

For me personally this issue is over. I took down the website of my own accord. I started the site to get Apple’s attention. Mission accomplished. Now that they have investigated what my site was alleging, and made their own statement and rectified the situation to my (and everyone who has a broken Nano screen) satisfaction, I have chosen to take down my site. Please do not ask me for pictures or emails from people with the problem. As this issue is resolved, I don’t think any of this information is relevant to anyone. I never started out to bash Apple and I am going to keep my integrity on that by removing the site and thanking Apple for looking into this and replacing the screens of those people who have been unfortunate to receive one of the “vendor problem” screens.

The whole Nano storm broke a couple of weeks ago, and had the potential to be a full blown blog/PR nightmare. Apple, unlike Dell, was paying attention. The question to ask yourself is this - would your marketing/PR group know what to do in a similar situation?

Via Scoble, I ran across two meme/search services I hadn't been aware of - Blogniscient (which is in somewhat the same space as memeorandum), and Digg (which looks like it's a cross between that space and straight blog search).

Right off the top, Blogniscient just fails the seriousness test for me. Why? It doesn't provide either RSS or Atom. Well heck - I don't track stuff via the browser anymore - I use an aggregator. Which is not to say that they should be XML only, but they sure as heck ought to be offering it. I don't have the time to hit bookmarks folks - get with the program :)

Digg is an interesting beast - it looks like a cross between memeorandum and feedster, and - like both - it offers syndication services. I was able to add Digg to my list of search providers in BottomFeeder quickly - their syndication feeds follow a nice simple format. I'll check back with Blogniscient from time to time, but until they offer syndication, they'll be mostly off my radar.

Looks like Patrick Logan is getting too much spam - he's turned his comments off. For the moment, I seem to have that problem mostly handled here, via three simple actions:

1. Any post that ages off the main view has comments turned off
2. Any post with more than N hrefs (it's configurable) is assumed to be spam
3. Comments that come in "too fast" get throttled out

That first one can be modified to allow a post's comments to stay on - in which case, an item specific feed is created, allowing tracking. All in all, it's worked pretty well. I get spam, but rarely. I suspect that being an "off brand" server helps as well.

Charles Miller explains, in copious detail:

Given the OPML spec, and the above examples, we can now ask ourselves a simple question: What is the difference between accepting OPML, and accepting arbitrary XML documents of unknown formats?

Answer: An OPML document limits where you can put text nodes.

Read the whole thing - it explains what I meant when I said I had to constantly munge my OPML import support in BottomFeeder early on. Even now, it works only because the entire field of aggregator developers has seems to have a simple attitude about that bit of code in their applications: "If I don't touch it, maybe everyone else will leave it alone too".

Word is, Office 12 will support PDF - both import and creation. Now that's a feature that's actually useful, unlike most of the damage that's been added to Office over the last few years. Still on their todo list - take the ugly stick away from whoever is wielding it against Outlook.

Go out and see it. Whedon wraps up the loose ends (that's as much of a spoiler as you'll get out of me), and it makes for some exciting cinema. Highly recommended.

The Yankees have won their 8th straight AL East championship - now, the perfect ending would be for them to knock the Red Sox out of the playoffs. Time to root for the Indians tomorrow :)

It's time for my weekly look at the server logs. The download rate for BottomFeeder is continuing to be strong - they ran at better than 650 per day last week. The distribution of downloads follows:

Platform	BottomFeeder Downloads
Windows	1927
HPUX	535
Sources	533
Mac 8/9	446
Update	277
Linux x86	231
Mac X	205
CE ARM	107
AIX	57
Solaris	56
Windows98/ME	51
SGI	44
Linux Sparc	34
ADUX	28
Linux PPC	28
CE x86	8
Source Script	5

Being the "off brand" choice for a number of platforms that aren't well served seems to help :) Next, let's see what things looked in in terms of tool access to the HTML pages.

Tool	Percentage of Accesses
Mozilla	51.8%
Internet Explorer	37.4%
Other	6%
Google Bot	2.5%
MSN Bot	1.3%
Opera	1%

Looks like my readers still prefer Mozilla based browsers. It also looks like Microsoft has their bot more properly tuned - it's not grabbing an absurd share in that list anymore. On to the RSS hits:

Tool	Percentage of Accesses
Mozilla	26.7%
BottomFeeder	17.6%
Other	12.9%
Net News Wire	10%
Safari RSS	4.1%
Planet Smalltalk	3.7%
NewsGator	3.2%
Magpie	2.9%
Internet Explorer	2.7%
Feed Demon	2.5%
BlogLines	2.3%
SharpReader	2.2%
Feed Reader	1.9%
RSS Bandit	1.7%
BlogSearch	1.3%
Liferea	1.3%
JetBrains	1%
Feed Tagger	1%
Google Bot	1%

I'm sure we'll see consolidation in the aggregator space eventually, but it's sure not happening yet - that tool distribution is still pretty varied. It's especially interesting to me how much - nearly 13% - falls into "other" (i.e., nothing in there individually approaches 1%). The field is still wide open.